Assessment of Telestroke Capacity in US Hospitals

Reperfusion via intravenous alteplase or mechanical thrombectomy reduces the damage of an acute ischemic stroke, but these treatments must be administered within a narrow time frame after onset of symptoms and not all patients can safely receive reperfusion. Larger hospitals typically have on-call stroke teams that quickly assess patients with suspected stroke; patients receiving care at hospitals without this expertise may be less likely to receive appropriate treatment.¹ Telestroke has emerged as a potential solution by which remote stroke experts guide local emergency physicians through the neurological examination, review of imaging, and management decisions.¹

Limited evidence exists on the prevalence of telestroke capacity in hospitals.^2,3 There is no comprehensive survey of when hospitals introduced their telestroke programs, and the use of telestroke cannot be reliably identified in insurance claims.⁴ We describe the growth of telestroke capacity in US hospitals and compare the characteristics of the hospitals with and without telestroke capacity.

In consultation with stroke experts about existing telestroke programs in the US and review of the literature, we identified and contacted 15 health care systems and private companies with known telestroke capacity (Figure). All agreed to participate and provided the names and contract start dates of hospitals in their networks on the requesting end of telestroke consults (spoke hospitals). This is a convenience sample and does not capture all telestroke sites and most likely underrepresents smaller networks. Harvard Medical School waived insitutional review board approval for this study because the study did not involve human subjects data.

Our sample included all short-term acute care hospitals and critical access hospitals open in 2017 (using most recent data available) with at least 1 stroke hospitalization in Medicare fee-for-service data. We linked our list of hospitals with telestroke capacity to this list using hospital name and city.

We described telestroke uptake from January 2001 to December 2019 and then compared characteristics of hospitals with telestroke capacity by 2019 vs those without via χ² tests. Hospital characteristics came from 2017 Medicare data for hospital rural/urban status, census region, and bed count, 2016 Centers for Disease Control and Prevention data for neurologists per capita ratio by hospital county, and 2017 American Hospital Association survey data for all other characteristics (Table). P values were 2-sided, and the signifiance threshold was .05.

By 2019, 1306 of 4751 hospitals (27.4%) had telestroke capacity. Uptake grew slowly and peaked in 2014 (Figure). Of hospitals open in 2017, hospitals with telestroke capacity were larger (523 [33.5%] in the highest quartile with telestroke capacity vs 201 [14.8%] in the lowest; P < .001; χ² = 212.1), more likely to be short-term acute care hospitals (1077 [31.7%] vs 229 [17.0%] of critical-access hospitals; P < .001; χ² = 105.2), and in urban areas (856 [30.2%] vs 450 [23.5%] rural; P < .001; χ² = 25.8). They were also more likely to be in the West, in counties with fewer neurologists per capita, have for-profit ownership, and have intensive care units (Table).

In one of the first national compilations of telestroke capacity to our knowledge, we found that approximately a quarter of hospitals now have access to this technology for patients in their emergency departments. Given that telestroke capacity is associated with greater use of reperfusion,⁵ the growing use of telestroke in the US hopefully leads to more reperfusion and improved outcomes.

Smaller, rural, and critical access hospitals, the group of hospitals that might benefit most from telestroke,⁶ are relatively less likely to have adopted this technology. It is unclear what barriers these hospitals face in introducing telestroke capacity. Potential barriers include lack of capital investment, lack of high-speed internet in rural communities, and regulatory barriers (eg, Stark Laws limit the material support that can be provided to referring hospitals).

The key limitation to this research is that our sample of telestroke networks was identified primarily through consultations with experts and as such may overrepresent larger systems of care anchored by teaching hospitals or using private telestroke companies.

Corresponding Author: Ateev Mehrotra, MD, MPH, Department of Health Care Policy, Harvard Medical School, 180 Longwood Ave, Boston, MA 02115 (mehrotra@hcp.med.harvard.edu).

Published Online: May 26, 2020. doi:10.1001/jamaneurol.2020.1274

Author Contributions: Dr Mehrotra had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Richard, Uscher-Pines, Zachrison, Siddiqui, Mehrotra.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Richard.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Richard, Zachrison, Mehrotra.

Obtained funding: Uscher-Pines, Mehrotra.

Administrative, technical, or material support: Richard, Wilcock.

Supervision: Schwamm, Uscher-Pines, Mehrotra.

Conflict of Interest Disclosures: Dr Schwamm reports serving as a scientific consultant to Genentech regarding trial design and conduct on late window thrombolysis, to Life Image regarding user interface design and usability, and to the Massachusetts Department of Public Health on stroke systems of care; reports grant funding from Comparative Effectiveness of Early Integrated Telehealth Versus In-Person Palliative Care for Patients with Advanced Lung Cancer REACH-PC and Patient-Centered Outcomes Research Institute; and serves as a member of the international advisory board for The Lancet-Digital Health (2019-present). Dr Zachrison reports grant funding from the Agency for Healthcare Research and Quality and the American Heart Association. No other disclosures were reported.

Funding/Support: This research was supported by the National Institute of Neurological Disorders and Stroke (grant R01NS111952).

Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.